Roughly one million Americans with inadequately controlled severe persistent asthma have a particularly high risk of exacerbations, hospitalizations, and death, and account for a per-patient health care burden exceeding $12,000 annually. Bronchial thermoplasty (BT) applies radiofrequency heating to airway smooth muscle reducing its thickness and hyper-reactivity. This novel minimally invasive intervention is applied to all airways 3mm and larger over the course of three procedures. We hypothesize that hyperpolarized xenon (HXe) MRI could identify a subset of dysfunctional airways that dominate the patient's symptoms, and that an image-guided single-session procedure could eliminate overtreatment of non-involved airways, minimize risk and lost productivity, reduce costs of managing this patient group, and convince more payers to reimburse. We propose a double-blind pilot study of thirty patients with severe asthma electing to undergo BT. All patients will participate in three HXe MRI pulmonary functional imaging evaluations of their ventilation defects: at baseline, after their first BT treatment and recovery, and after their final BT treatmet and recovery. Existing and new software methods will identify and rank problematic segmental airways by linking them with the observed ventilation defects. While all patients will receive the full course of BT treatment, half of them will receive a guided BT treatment targeting the most problematic airways in their first treatment session while the control group will receive treatment following the standard sequence. With thirty subjects, this pilot study is adequately powered to detect a change in Quality of Life (QoL) from baseline after one image-guided BT treatment, if changes are similar to published BT trials. We seek, however, greater understanding of biopredictors of BT response from our data. We hypothesize that HXe MRI will provide a metric for grading asthma severity more quantitative than QoL for evaluating BT. An important aim of this study is to evaluate this HXe MRI biomarker for asthma disease severity that includes a set of four images that determine average airway obstruction, bronchodilator reactivity, and variation of these metrics with time. With the HXe MRI metric as our primary endpoint and several asthma QoL indices as secondary end-points, we compare the single session image-guided BT treatment to the standard three-session BT treatment after unblinding the study. Two comparisons are performed: the target group after the single-session guided BT treatment will be compared with the control group after the full three sessions; and will be compared with their own improvement after three sessions, serving as their own controls. If data support the most favorable outcome, that HXe MRI guided BT treatment can reduce overtreatment by two-thirds, cut BT costs in half, and achieve an equivalent benefit, then one million Americans could elect image-guided BT to improve their management of severe asthma. Finally, as a clinically-indicated companion diagnostic for an image-guided therapy, HXe would become an FDA-approved and commercially available diagnostic agent within a few years.